SlideShare una empresa de Scribd logo
1 de 12
Descargar para leer sin conexión
1
indoo.rs CaLibre
Unsupervised Database Free RSSI Calibration Indoor Navigation
Boxian Dong (indoo.rs, TU Wien), Thomas Burgess (indoo.rs), Hans-Berndt Neuner (TU Wien)
ENC 2017, Lausanne, May, 2017
CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017
indoo.rs GmbH.
2
Positioning &
Navigation
Asset Tracking
Analytics
CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017
➔ indoo.rs localization relies on RSSI fingerprint maps
➔ SLAM Engine (SE)
◆ Generates initial radio maps
◆ Uses radio data from dedicated recordings
➔ SLAM Crowd Engine (SCE)
◆ Updates, improve and expand radio maps
◆ Uses radio data crowd sourced from navigating users
indoo.rs SLAM crowd engine™.
3
Radio map
Update
Error estimation
CaLibre
SLAM
Localization
CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017
➔ RSSI must be made comparable
◆ RSSI only indicates relative power level
◆ RSSI characteristics differ between devices
◆ Crowd data combines many different devices
➔ CaLibre is our approach to this problem
◆ Calibrate all inputs in every job
● No need for a hard to maintain database
◆ Form tiles by grouping scans close to each other
● Scans in each group should have same power level
◆ Correlate differences all single tiles
● Create set of comparable samples
◆ Find linear relations from comparable samples
RSSI calibration.
4
CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017
CaLibre: signal preprocessing.
signal preprocessing
Input recordings
building radio
maps
grouping scans
calibration sample
creation
offset and slope
regression
calibration
results
denoised signal
grouping index
scanned signal
building
statistic
comparable RSSIs
High Noise ± 10dB
Missing signals
Sliding window average
based noise reduction
5
CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017
CaLibre: grouping scans in tiles.
signal preprocessing
Input recordings
building radio
maps
grouping scans
calibration sample
creation
offset and slope
regression
calibration
results
denoised signal
grouping index
scanned signal
building
statistic
comparable RSSIs
Group scans by
➔ Overlapping networks
➔ Having higher power than a
relative power threshold
(part of total visible area)
6
CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017
CaLibre: grouping scans in tiles.
signal preprocessing
Input recordings
building radio
maps
grouping scans
calibration sample
creation
offset and slope
regression
calibration
results
denoised signal
grouping index
scanned signal
building
statistic
comparable RSSIs
➔ Low network overlap
➔ Low power threshold
➔ High network overlap
➔ High power threshold
7
CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017
CaLibre: calibration sample creation.
signal preprocessing
Input recordings
building radio
maps
grouping scans
calibration sample
creation
offset and slope
regression
calibration
results
denoised signal
grouping index
scanned signal
building
statistic
comparable RSSIs
➔ High Noise
➔ Large fading effects
(bluetooth signal)
➔ Compute median and weight
8
CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017
CaLibre: regression.
signal preprocessing
Input recordings
building radio
maps
grouping scans
calibration sample
creation
offset and slope
regression
calibration
results
denoised signal
grouping index
scanned signal
building
statistic
comparable RSSIs
➔ One sample per network per tile
➔ Using weighted Ridge regression
◆ Computationally cheap while robust to noise
➔ Linear fit
◆ a - slope a
◆ Δ80
- offset at -80dB
9
CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017
CaLibre: test results.
10
➔ Calibrate multiple 1~2 minutes recordings
➔ Summarize statistically calibration results between
same pair devices
➔ Comparing Calibre results with manual results
CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017
CONCLUSIONS:
➔ Calibrate RSSI reading between recordings
➔ Tiling parameters depend on radio
environments
➔ Less efficient in recovery Slope
CaLibre: conclusions.
11
OUTLOOK:
➔ Calibration between recording and radio
map
➔ Tiling parameter further optimization
➔ Multiple recording calibration
CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017
Boxian Dong
Senior researcher
12

Más contenido relacionado

La actualidad más candente

Digimap Aerial Collection - Geoforum 2016 - Ian Holmes
Digimap Aerial Collection - Geoforum 2016 - Ian HolmesDigimap Aerial Collection - Geoforum 2016 - Ian Holmes
Digimap Aerial Collection - Geoforum 2016 - Ian HolmesEDINA, University of Edinburgh
 
Recommendation engine using Aerospike and/OR MongoDB
Recommendation engine using Aerospike and/OR MongoDBRecommendation engine using Aerospike and/OR MongoDB
Recommendation engine using Aerospike and/OR MongoDBPeter Milne
 
Review of Cloud Computing Simulation Platforms and Related Environments
Review of Cloud Computing Simulation Platforms and Related EnvironmentsReview of Cloud Computing Simulation Platforms and Related Environments
Review of Cloud Computing Simulation Platforms and Related EnvironmentsRECAP Project
 
RIPE Atlas and IXPs "Stitchin' it up"
RIPE Atlas and IXPs "Stitchin' it up"RIPE Atlas and IXPs "Stitchin' it up"
RIPE Atlas and IXPs "Stitchin' it up"RIPE NCC
 
FR3.TO5.5.pptx
FR3.TO5.5.pptxFR3.TO5.5.pptx
FR3.TO5.5.pptxgrssieee
 
Introduction to STILT – an on-demand CO2 footprint calculator service
Introduction to STILT – an on-demand CO2 footprint calculator serviceIntroduction to STILT – an on-demand CO2 footprint calculator service
Introduction to STILT – an on-demand CO2 footprint calculator serviceEUDAT
 
IoT Event Processing and Analytics with InfluxDB in Google Cloud | Christoph ...
IoT Event Processing and Analytics with InfluxDB in Google Cloud | Christoph ...IoT Event Processing and Analytics with InfluxDB in Google Cloud | Christoph ...
IoT Event Processing and Analytics with InfluxDB in Google Cloud | Christoph ...InfluxData
 
Are Dutch Internet Paths Local - A Measurement Study Using RIPE Atlas
Are Dutch Internet Paths Local - A Measurement Study Using RIPE AtlasAre Dutch Internet Paths Local - A Measurement Study Using RIPE Atlas
Are Dutch Internet Paths Local - A Measurement Study Using RIPE AtlasRIPE NCC
 
IXP Traffic and Major Sports Events
IXP Traffic and Major Sports EventsIXP Traffic and Major Sports Events
IXP Traffic and Major Sports EventsRIPE NCC
 
Deep Learning - Optimization Basic
Deep Learning - Optimization BasicDeep Learning - Optimization Basic
Deep Learning - Optimization BasicJaehyun Jun
 
The RIPE NCC, Internet Measurements and IXPs
The RIPE NCC, Internet Measurements and IXPsThe RIPE NCC, Internet Measurements and IXPs
The RIPE NCC, Internet Measurements and IXPsRIPE NCC
 
RIPE Atlas
RIPE AtlasRIPE Atlas
RIPE AtlasRIPE NCC
 
Architecting R into Storm Application Development Process
Architecting R into Storm Application Development ProcessArchitecting R into Storm Application Development Process
Architecting R into Storm Application Development ProcessDataWorks Summit
 
PEPS: CNES Sentinel Satellite Image Analysis, On-Premises and in the Cloud wi...
PEPS: CNES Sentinel Satellite Image Analysis, On-Premises and in the Cloud wi...PEPS: CNES Sentinel Satellite Image Analysis, On-Premises and in the Cloud wi...
PEPS: CNES Sentinel Satellite Image Analysis, On-Premises and in the Cloud wi...OW2
 
Large Scale Stream Analytics using a Resource-constrained Edge
Large Scale Stream Analytics using a Resource-constrained EdgeLarge Scale Stream Analytics using a Resource-constrained Edge
Large Scale Stream Analytics using a Resource-constrained EdgeGabriele Di Bernardo
 
Countries, IXPs and RIPE Atlas
Countries, IXPs and RIPE AtlasCountries, IXPs and RIPE Atlas
Countries, IXPs and RIPE AtlasRIPE NCC
 
Mapping K-PowerNet’s Fiber Optic Network
Mapping K-PowerNet’s Fiber Optic NetworkMapping K-PowerNet’s Fiber Optic Network
Mapping K-PowerNet’s Fiber Optic NetworkScott Kusselson
 
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...Seldon
 

La actualidad más candente (20)

Digimap Aerial Collection - Geoforum 2016 - Ian Holmes
Digimap Aerial Collection - Geoforum 2016 - Ian HolmesDigimap Aerial Collection - Geoforum 2016 - Ian Holmes
Digimap Aerial Collection - Geoforum 2016 - Ian Holmes
 
Recommendation engine using Aerospike and/OR MongoDB
Recommendation engine using Aerospike and/OR MongoDBRecommendation engine using Aerospike and/OR MongoDB
Recommendation engine using Aerospike and/OR MongoDB
 
Review of Cloud Computing Simulation Platforms and Related Environments
Review of Cloud Computing Simulation Platforms and Related EnvironmentsReview of Cloud Computing Simulation Platforms and Related Environments
Review of Cloud Computing Simulation Platforms and Related Environments
 
RIPE Atlas and IXPs "Stitchin' it up"
RIPE Atlas and IXPs "Stitchin' it up"RIPE Atlas and IXPs "Stitchin' it up"
RIPE Atlas and IXPs "Stitchin' it up"
 
FR3.TO5.5.pptx
FR3.TO5.5.pptxFR3.TO5.5.pptx
FR3.TO5.5.pptx
 
Introduction to STILT – an on-demand CO2 footprint calculator service
Introduction to STILT – an on-demand CO2 footprint calculator serviceIntroduction to STILT – an on-demand CO2 footprint calculator service
Introduction to STILT – an on-demand CO2 footprint calculator service
 
IoT Event Processing and Analytics with InfluxDB in Google Cloud | Christoph ...
IoT Event Processing and Analytics with InfluxDB in Google Cloud | Christoph ...IoT Event Processing and Analytics with InfluxDB in Google Cloud | Christoph ...
IoT Event Processing and Analytics with InfluxDB in Google Cloud | Christoph ...
 
Are Dutch Internet Paths Local - A Measurement Study Using RIPE Atlas
Are Dutch Internet Paths Local - A Measurement Study Using RIPE AtlasAre Dutch Internet Paths Local - A Measurement Study Using RIPE Atlas
Are Dutch Internet Paths Local - A Measurement Study Using RIPE Atlas
 
IXP Traffic and Major Sports Events
IXP Traffic and Major Sports EventsIXP Traffic and Major Sports Events
IXP Traffic and Major Sports Events
 
Deep Learning - Optimization Basic
Deep Learning - Optimization BasicDeep Learning - Optimization Basic
Deep Learning - Optimization Basic
 
Geoforum - Digimap Update 2015
Geoforum - Digimap Update 2015Geoforum - Digimap Update 2015
Geoforum - Digimap Update 2015
 
The RIPE NCC, Internet Measurements and IXPs
The RIPE NCC, Internet Measurements and IXPsThe RIPE NCC, Internet Measurements and IXPs
The RIPE NCC, Internet Measurements and IXPs
 
RIPE Atlas
RIPE AtlasRIPE Atlas
RIPE Atlas
 
Architecting R into Storm Application Development Process
Architecting R into Storm Application Development ProcessArchitecting R into Storm Application Development Process
Architecting R into Storm Application Development Process
 
PEPS: CNES Sentinel Satellite Image Analysis, On-Premises and in the Cloud wi...
PEPS: CNES Sentinel Satellite Image Analysis, On-Premises and in the Cloud wi...PEPS: CNES Sentinel Satellite Image Analysis, On-Premises and in the Cloud wi...
PEPS: CNES Sentinel Satellite Image Analysis, On-Premises and in the Cloud wi...
 
Large Scale Stream Analytics using a Resource-constrained Edge
Large Scale Stream Analytics using a Resource-constrained EdgeLarge Scale Stream Analytics using a Resource-constrained Edge
Large Scale Stream Analytics using a Resource-constrained Edge
 
Countries, IXPs and RIPE Atlas
Countries, IXPs and RIPE AtlasCountries, IXPs and RIPE Atlas
Countries, IXPs and RIPE Atlas
 
Mapping K-PowerNet’s Fiber Optic Network
Mapping K-PowerNet’s Fiber Optic NetworkMapping K-PowerNet’s Fiber Optic Network
Mapping K-PowerNet’s Fiber Optic Network
 
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...
Tensorflow London 13: Barbara Fusinska 'Hassle Free, Scalable, Machine Learni...
 
Archive concept
Archive conceptArchive concept
Archive concept
 

Similar a Unsupervised Database Free RSSI Calibration Indoor Navigation

Dense Retrieval with Apache Solr Neural Search.pdf
Dense Retrieval with Apache Solr Neural Search.pdfDense Retrieval with Apache Solr Neural Search.pdf
Dense Retrieval with Apache Solr Neural Search.pdfSease
 
Geospatial Sensor Networks and Partitioning Data
Geospatial Sensor Networks and Partitioning DataGeospatial Sensor Networks and Partitioning Data
Geospatial Sensor Networks and Partitioning DataAlexMiowski
 
Launching Your First Big Data Project on AWS
Launching Your First Big Data Project on AWSLaunching Your First Big Data Project on AWS
Launching Your First Big Data Project on AWSAmazon Web Services
 
A Graph Database That Scales - ArangoDB 3.7 Release Webinar
A Graph Database That Scales - ArangoDB 3.7 Release WebinarA Graph Database That Scales - ArangoDB 3.7 Release Webinar
A Graph Database That Scales - ArangoDB 3.7 Release WebinarArangoDB Database
 
Scaling Open Source Big Data Cloud Applications is Easy/Hard
Scaling Open Source Big Data Cloud Applications is Easy/HardScaling Open Source Big Data Cloud Applications is Easy/Hard
Scaling Open Source Big Data Cloud Applications is Easy/HardPaul Brebner
 
Vectorized Deep Learning Acceleration from Preprocessing to Inference and Tra...
Vectorized Deep Learning Acceleration from Preprocessing to Inference and Tra...Vectorized Deep Learning Acceleration from Preprocessing to Inference and Tra...
Vectorized Deep Learning Acceleration from Preprocessing to Inference and Tra...Databricks
 
Flink Forward SF 2017: Malo Deniélou - No shard left behind: Dynamic work re...
Flink Forward SF 2017: Malo Deniélou -  No shard left behind: Dynamic work re...Flink Forward SF 2017: Malo Deniélou -  No shard left behind: Dynamic work re...
Flink Forward SF 2017: Malo Deniélou - No shard left behind: Dynamic work re...Flink Forward
 
Big Data Meets NVM: Accelerating Big Data Processing with Non-Volatile Memory...
Big Data Meets NVM: Accelerating Big Data Processing with Non-Volatile Memory...Big Data Meets NVM: Accelerating Big Data Processing with Non-Volatile Memory...
Big Data Meets NVM: Accelerating Big Data Processing with Non-Volatile Memory...DataWorks Summit
 
AWS re:Invent 2016: Large-Scale, Cloud-Based Analysis of Cancer Genomes: Less...
AWS re:Invent 2016: Large-Scale, Cloud-Based Analysis of Cancer Genomes: Less...AWS re:Invent 2016: Large-Scale, Cloud-Based Analysis of Cancer Genomes: Less...
AWS re:Invent 2016: Large-Scale, Cloud-Based Analysis of Cancer Genomes: Less...Amazon Web Services
 
Malo Denielou - No shard left behind: Dynamic work rebalancing in Apache Beam
Malo Denielou - No shard left behind: Dynamic work rebalancing in Apache BeamMalo Denielou - No shard left behind: Dynamic work rebalancing in Apache Beam
Malo Denielou - No shard left behind: Dynamic work rebalancing in Apache BeamFlink Forward
 
AWS를 활용한 Big Data 실전 배치 사례 :: 이한주 :: AWS Summit Seoul 2016
AWS를 활용한 Big Data 실전 배치 사례 :: 이한주 :: AWS Summit Seoul 2016AWS를 활용한 Big Data 실전 배치 사례 :: 이한주 :: AWS Summit Seoul 2016
AWS를 활용한 Big Data 실전 배치 사례 :: 이한주 :: AWS Summit Seoul 2016Amazon Web Services Korea
 
Neural Search Comes to Apache Solr_ Approximate Nearest Neighbor, BERT and Mo...
Neural Search Comes to Apache Solr_ Approximate Nearest Neighbor, BERT and Mo...Neural Search Comes to Apache Solr_ Approximate Nearest Neighbor, BERT and Mo...
Neural Search Comes to Apache Solr_ Approximate Nearest Neighbor, BERT and Mo...Sease
 
Real-Time Vote Platform Benchmark
Real-Time Vote Platform BenchmarkReal-Time Vote Platform Benchmark
Real-Time Vote Platform BenchmarkLahav Savir
 
Streaming data for real time analysis
Streaming data for real time analysisStreaming data for real time analysis
Streaming data for real time analysisAmazon Web Services
 
찾아가는 AWS 세미나(구로,가산,판교) - AWS 기반 빅데이터 활용 방법 (김일호 솔루션즈 아키텍트)
찾아가는 AWS 세미나(구로,가산,판교) - AWS 기반 빅데이터 활용 방법 (김일호 솔루션즈 아키텍트)찾아가는 AWS 세미나(구로,가산,판교) - AWS 기반 빅데이터 활용 방법 (김일호 솔루션즈 아키텍트)
찾아가는 AWS 세미나(구로,가산,판교) - AWS 기반 빅데이터 활용 방법 (김일호 솔루션즈 아키텍트)Amazon Web Services Korea
 
BDE SC3.3 Workshop - Data management in WT testing and monitoring
 BDE SC3.3 Workshop - Data management in WT testing and monitoring  BDE SC3.3 Workshop - Data management in WT testing and monitoring
BDE SC3.3 Workshop - Data management in WT testing and monitoring BigData_Europe
 
Empowering the AWS DynamoDB™ application developer with Alternator
Empowering the AWS DynamoDB™ application developer with AlternatorEmpowering the AWS DynamoDB™ application developer with Alternator
Empowering the AWS DynamoDB™ application developer with AlternatorScyllaDB
 
Deview 2013 rise of the wimpy machines - john mao
Deview 2013   rise of the wimpy machines - john maoDeview 2013   rise of the wimpy machines - john mao
Deview 2013 rise of the wimpy machines - john maoNAVER D2
 
Best Practices for Building Open Source Data Layers
Best Practices for Building Open Source Data LayersBest Practices for Building Open Source Data Layers
Best Practices for Building Open Source Data LayersIBMCompose
 
SDVIs and In-Situ Visualization on TACC's Stampede
SDVIs and In-Situ Visualization on TACC's StampedeSDVIs and In-Situ Visualization on TACC's Stampede
SDVIs and In-Situ Visualization on TACC's StampedeIntel® Software
 

Similar a Unsupervised Database Free RSSI Calibration Indoor Navigation (20)

Dense Retrieval with Apache Solr Neural Search.pdf
Dense Retrieval with Apache Solr Neural Search.pdfDense Retrieval with Apache Solr Neural Search.pdf
Dense Retrieval with Apache Solr Neural Search.pdf
 
Geospatial Sensor Networks and Partitioning Data
Geospatial Sensor Networks and Partitioning DataGeospatial Sensor Networks and Partitioning Data
Geospatial Sensor Networks and Partitioning Data
 
Launching Your First Big Data Project on AWS
Launching Your First Big Data Project on AWSLaunching Your First Big Data Project on AWS
Launching Your First Big Data Project on AWS
 
A Graph Database That Scales - ArangoDB 3.7 Release Webinar
A Graph Database That Scales - ArangoDB 3.7 Release WebinarA Graph Database That Scales - ArangoDB 3.7 Release Webinar
A Graph Database That Scales - ArangoDB 3.7 Release Webinar
 
Scaling Open Source Big Data Cloud Applications is Easy/Hard
Scaling Open Source Big Data Cloud Applications is Easy/HardScaling Open Source Big Data Cloud Applications is Easy/Hard
Scaling Open Source Big Data Cloud Applications is Easy/Hard
 
Vectorized Deep Learning Acceleration from Preprocessing to Inference and Tra...
Vectorized Deep Learning Acceleration from Preprocessing to Inference and Tra...Vectorized Deep Learning Acceleration from Preprocessing to Inference and Tra...
Vectorized Deep Learning Acceleration from Preprocessing to Inference and Tra...
 
Flink Forward SF 2017: Malo Deniélou - No shard left behind: Dynamic work re...
Flink Forward SF 2017: Malo Deniélou -  No shard left behind: Dynamic work re...Flink Forward SF 2017: Malo Deniélou -  No shard left behind: Dynamic work re...
Flink Forward SF 2017: Malo Deniélou - No shard left behind: Dynamic work re...
 
Big Data Meets NVM: Accelerating Big Data Processing with Non-Volatile Memory...
Big Data Meets NVM: Accelerating Big Data Processing with Non-Volatile Memory...Big Data Meets NVM: Accelerating Big Data Processing with Non-Volatile Memory...
Big Data Meets NVM: Accelerating Big Data Processing with Non-Volatile Memory...
 
AWS re:Invent 2016: Large-Scale, Cloud-Based Analysis of Cancer Genomes: Less...
AWS re:Invent 2016: Large-Scale, Cloud-Based Analysis of Cancer Genomes: Less...AWS re:Invent 2016: Large-Scale, Cloud-Based Analysis of Cancer Genomes: Less...
AWS re:Invent 2016: Large-Scale, Cloud-Based Analysis of Cancer Genomes: Less...
 
Malo Denielou - No shard left behind: Dynamic work rebalancing in Apache Beam
Malo Denielou - No shard left behind: Dynamic work rebalancing in Apache BeamMalo Denielou - No shard left behind: Dynamic work rebalancing in Apache Beam
Malo Denielou - No shard left behind: Dynamic work rebalancing in Apache Beam
 
AWS를 활용한 Big Data 실전 배치 사례 :: 이한주 :: AWS Summit Seoul 2016
AWS를 활용한 Big Data 실전 배치 사례 :: 이한주 :: AWS Summit Seoul 2016AWS를 활용한 Big Data 실전 배치 사례 :: 이한주 :: AWS Summit Seoul 2016
AWS를 활용한 Big Data 실전 배치 사례 :: 이한주 :: AWS Summit Seoul 2016
 
Neural Search Comes to Apache Solr_ Approximate Nearest Neighbor, BERT and Mo...
Neural Search Comes to Apache Solr_ Approximate Nearest Neighbor, BERT and Mo...Neural Search Comes to Apache Solr_ Approximate Nearest Neighbor, BERT and Mo...
Neural Search Comes to Apache Solr_ Approximate Nearest Neighbor, BERT and Mo...
 
Real-Time Vote Platform Benchmark
Real-Time Vote Platform BenchmarkReal-Time Vote Platform Benchmark
Real-Time Vote Platform Benchmark
 
Streaming data for real time analysis
Streaming data for real time analysisStreaming data for real time analysis
Streaming data for real time analysis
 
찾아가는 AWS 세미나(구로,가산,판교) - AWS 기반 빅데이터 활용 방법 (김일호 솔루션즈 아키텍트)
찾아가는 AWS 세미나(구로,가산,판교) - AWS 기반 빅데이터 활용 방법 (김일호 솔루션즈 아키텍트)찾아가는 AWS 세미나(구로,가산,판교) - AWS 기반 빅데이터 활용 방법 (김일호 솔루션즈 아키텍트)
찾아가는 AWS 세미나(구로,가산,판교) - AWS 기반 빅데이터 활용 방법 (김일호 솔루션즈 아키텍트)
 
BDE SC3.3 Workshop - Data management in WT testing and monitoring
 BDE SC3.3 Workshop - Data management in WT testing and monitoring  BDE SC3.3 Workshop - Data management in WT testing and monitoring
BDE SC3.3 Workshop - Data management in WT testing and monitoring
 
Empowering the AWS DynamoDB™ application developer with Alternator
Empowering the AWS DynamoDB™ application developer with AlternatorEmpowering the AWS DynamoDB™ application developer with Alternator
Empowering the AWS DynamoDB™ application developer with Alternator
 
Deview 2013 rise of the wimpy machines - john mao
Deview 2013   rise of the wimpy machines - john maoDeview 2013   rise of the wimpy machines - john mao
Deview 2013 rise of the wimpy machines - john mao
 
Best Practices for Building Open Source Data Layers
Best Practices for Building Open Source Data LayersBest Practices for Building Open Source Data Layers
Best Practices for Building Open Source Data Layers
 
SDVIs and In-Situ Visualization on TACC's Stampede
SDVIs and In-Situ Visualization on TACC's StampedeSDVIs and In-Situ Visualization on TACC's Stampede
SDVIs and In-Situ Visualization on TACC's Stampede
 

Último

detection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptxdetection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptxAleenaJamil4
 
Real-Time AI Streaming - AI Max Princeton
Real-Time AI  Streaming - AI Max PrincetonReal-Time AI  Streaming - AI Max Princeton
Real-Time AI Streaming - AI Max PrincetonTimothy Spann
 
Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...
Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...
Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...Boston Institute of Analytics
 
RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.natarajan8993
 
Student profile product demonstration on grades, ability, well-being and mind...
Student profile product demonstration on grades, ability, well-being and mind...Student profile product demonstration on grades, ability, well-being and mind...
Student profile product demonstration on grades, ability, well-being and mind...Seán Kennedy
 
Identifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population MeanIdentifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population MeanMYRABACSAFRA2
 
Heart Disease Classification Report: A Data Analysis Project
Heart Disease Classification Report: A Data Analysis ProjectHeart Disease Classification Report: A Data Analysis Project
Heart Disease Classification Report: A Data Analysis ProjectBoston Institute of Analytics
 
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDINTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDRafezzaman
 
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...Amil Baba Dawood bangali
 
Multiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfMultiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfchwongval
 
Generative AI for Social Good at Open Data Science East 2024
Generative AI for Social Good at Open Data Science East 2024Generative AI for Social Good at Open Data Science East 2024
Generative AI for Social Good at Open Data Science East 2024Colleen Farrelly
 
Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)Cathrine Wilhelmsen
 
Top 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In QueensTop 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In Queensdataanalyticsqueen03
 
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024Timothy Spann
 
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default  Presentation : Data Analysis Project PPTPredictive Analysis for Loan Default  Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPTBoston Institute of Analytics
 
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...limedy534
 
ASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel CanterASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel Cantervoginip
 
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改yuu sss
 
Learn How Data Science Changes Our World
Learn How Data Science Changes Our WorldLearn How Data Science Changes Our World
Learn How Data Science Changes Our WorldEduminds Learning
 

Último (20)

detection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptxdetection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptx
 
Real-Time AI Streaming - AI Max Princeton
Real-Time AI  Streaming - AI Max PrincetonReal-Time AI  Streaming - AI Max Princeton
Real-Time AI Streaming - AI Max Princeton
 
Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...
Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...
Decoding the Heart: Student Presentation on Heart Attack Prediction with Data...
 
RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.
 
Student profile product demonstration on grades, ability, well-being and mind...
Student profile product demonstration on grades, ability, well-being and mind...Student profile product demonstration on grades, ability, well-being and mind...
Student profile product demonstration on grades, ability, well-being and mind...
 
Identifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population MeanIdentifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population Mean
 
Heart Disease Classification Report: A Data Analysis Project
Heart Disease Classification Report: A Data Analysis ProjectHeart Disease Classification Report: A Data Analysis Project
Heart Disease Classification Report: A Data Analysis Project
 
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTDINTERNSHIP ON PURBASHA COMPOSITE TEX LTD
INTERNSHIP ON PURBASHA COMPOSITE TEX LTD
 
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
 
Multiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfMultiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdf
 
Generative AI for Social Good at Open Data Science East 2024
Generative AI for Social Good at Open Data Science East 2024Generative AI for Social Good at Open Data Science East 2024
Generative AI for Social Good at Open Data Science East 2024
 
Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)
 
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
 
Top 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In QueensTop 5 Best Data Analytics Courses In Queens
Top 5 Best Data Analytics Courses In Queens
 
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
 
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default  Presentation : Data Analysis Project PPTPredictive Analysis for Loan Default  Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPT
 
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...
 
ASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel CanterASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel Canter
 
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
专业一比一美国俄亥俄大学毕业证成绩单pdf电子版制作修改
 
Learn How Data Science Changes Our World
Learn How Data Science Changes Our WorldLearn How Data Science Changes Our World
Learn How Data Science Changes Our World
 

Unsupervised Database Free RSSI Calibration Indoor Navigation

  • 1. 1 indoo.rs CaLibre Unsupervised Database Free RSSI Calibration Indoor Navigation Boxian Dong (indoo.rs, TU Wien), Thomas Burgess (indoo.rs), Hans-Berndt Neuner (TU Wien) ENC 2017, Lausanne, May, 2017
  • 2. CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017 indoo.rs GmbH. 2 Positioning & Navigation Asset Tracking Analytics
  • 3. CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017 ➔ indoo.rs localization relies on RSSI fingerprint maps ➔ SLAM Engine (SE) ◆ Generates initial radio maps ◆ Uses radio data from dedicated recordings ➔ SLAM Crowd Engine (SCE) ◆ Updates, improve and expand radio maps ◆ Uses radio data crowd sourced from navigating users indoo.rs SLAM crowd engine™. 3 Radio map Update Error estimation CaLibre SLAM Localization
  • 4. CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017 ➔ RSSI must be made comparable ◆ RSSI only indicates relative power level ◆ RSSI characteristics differ between devices ◆ Crowd data combines many different devices ➔ CaLibre is our approach to this problem ◆ Calibrate all inputs in every job ● No need for a hard to maintain database ◆ Form tiles by grouping scans close to each other ● Scans in each group should have same power level ◆ Correlate differences all single tiles ● Create set of comparable samples ◆ Find linear relations from comparable samples RSSI calibration. 4
  • 5. CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017 CaLibre: signal preprocessing. signal preprocessing Input recordings building radio maps grouping scans calibration sample creation offset and slope regression calibration results denoised signal grouping index scanned signal building statistic comparable RSSIs High Noise ± 10dB Missing signals Sliding window average based noise reduction 5
  • 6. CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017 CaLibre: grouping scans in tiles. signal preprocessing Input recordings building radio maps grouping scans calibration sample creation offset and slope regression calibration results denoised signal grouping index scanned signal building statistic comparable RSSIs Group scans by ➔ Overlapping networks ➔ Having higher power than a relative power threshold (part of total visible area) 6
  • 7. CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017 CaLibre: grouping scans in tiles. signal preprocessing Input recordings building radio maps grouping scans calibration sample creation offset and slope regression calibration results denoised signal grouping index scanned signal building statistic comparable RSSIs ➔ Low network overlap ➔ Low power threshold ➔ High network overlap ➔ High power threshold 7
  • 8. CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017 CaLibre: calibration sample creation. signal preprocessing Input recordings building radio maps grouping scans calibration sample creation offset and slope regression calibration results denoised signal grouping index scanned signal building statistic comparable RSSIs ➔ High Noise ➔ Large fading effects (bluetooth signal) ➔ Compute median and weight 8
  • 9. CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017 CaLibre: regression. signal preprocessing Input recordings building radio maps grouping scans calibration sample creation offset and slope regression calibration results denoised signal grouping index scanned signal building statistic comparable RSSIs ➔ One sample per network per tile ➔ Using weighted Ridge regression ◆ Computationally cheap while robust to noise ➔ Linear fit ◆ a - slope a ◆ Δ80 - offset at -80dB 9
  • 10. CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017 CaLibre: test results. 10 ➔ Calibrate multiple 1~2 minutes recordings ➔ Summarize statistically calibration results between same pair devices ➔ Comparing Calibre results with manual results
  • 11. CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017 CONCLUSIONS: ➔ Calibrate RSSI reading between recordings ➔ Tiling parameters depend on radio environments ➔ Less efficient in recovery Slope CaLibre: conclusions. 11 OUTLOOK: ➔ Calibration between recording and radio map ➔ Tiling parameter further optimization ➔ Multiple recording calibration
  • 12. CaLibre - Boxian Dong <boxian@indoo.rs> - ENC2017 Boxian Dong Senior researcher 12